some doubts in garch models

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

some doubts in garch models

Ricardo Zambrano Aguilera
Hi everyone i need some help , and a have a few questions....


i´m looking for an alghoritm to simulate garch process, (especially garch(1,1))
somebody can help me??


i need to know if our software has the Lagrange multiplier test of Engle to test for conditional heteroscedasticity??
saludos
ricardo

i need to Extract Standardized Residuals from a garch model, to test all the properties...it´necessary or i commit a mistake, because it´s already standarized??


to get the data into (for analize...) garch process, always their Kurtosis must be larger than 3??

thxs


greetings
ricardo zambrano ([hidden email])

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Reply | Threaded
Open this post in threaded view
|

Re: some doubts in garch models

Spencer Graves
          I just got 21 hits from RSiteSearch("GARCH", "function"), including
the following:
          http://finzi.psych.upenn.edu/R/library/fSeries/html/00Index.html

http://finzi.psych.upenn.edu/R/library/tseries/html/00Index.html

http://finzi.psych.upenn.edu/R/library/fOptions/html/00Index.html

          Si esto no está suficiently, favor de leer el posting guide!
"www.R-project.org/posting-guide.html".  Cuando no puedo yo entender
bien su pregunta, tengo dificutades en mis intentos para servirle.

          Spencer Graves

Ricardo Zambrano Aguilera wrote:

> Hi everyone i need some help , and a have a few questions....
>
>
> i´m looking for an alghoritm to simulate garch process, (especially garch(1,1))
> somebody can help me??
>
>
> i need to know if our software has the Lagrange multiplier test of Engle to test for conditional heteroscedasticity??
> saludos
> ricardo
>
> i need to Extract Standardized Residuals from a garch model, to test all the properties...it´necessary or i commit a mistake, because it´s already standarized??
>
>
> to get the data into (for analize...) garch process, always their Kurtosis must be larger than 3??
>
> thxs
>
>
> greetings
> ricardo zambrano ([hidden email])
>
> _______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Reply | Threaded
Open this post in threaded view
|

Re: some doubts in garch models

Diethelm Wuertz
In reply to this post by Ricardo Zambrano Aguilera
Ricardo Zambrano Aguilera wrote:

>Hi everyone i need some help , and a have a few questions....
>
>
>i´m looking for an alghoritm to simulate garch process, (especially garch(1,1))
>somebody can help me??
>  
>
The new garchFit() function in fSeries 221.10065 in Rmetrics can do this
job.
(It reproduces the FCP bechmark.)

>
>i need to know if our software has the Lagrange multiplier test of Engle to test for conditional heteroscedasticity??
>saludos
>ricardo
>  
>
Yes

>i need to Extract Standardized Residuals from a garch model, to test all the properties...it´necessary or i commit a mistake, because it´s already standarized??
>  
>
That is automatically be done

Have a look on the garch paper on www.rmetrics.org

best regards
DW

>
>to get the data into (for analize...) garch process, always their Kurtosis must be larger than 3??
>
>thxs
>
>
>greetings
>ricardo zambrano ([hidden email])
>
>_______________________________________________
>[hidden email] mailing list
>https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>
>  
>

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Reply | Threaded
Open this post in threaded view
|

Re: some doubts in garch models

Ricardo Zambrano Aguilera
In reply to this post by Ricardo Zambrano Aguilera

        Thank a lot mr Diethelm Wuertz and mr Patrick Burns my best regards..., but then i have another doubt... and it is a big
        if my data... the exchange rate (dollar (USA) - peso(Chile)), between 1-1-2001 - 12/12/2005, (i´m working with the log-returns), my first intention is to aplly an ARMA -GARCH , but  their Kurtosis is 1.182 < 3,  ¿ can i use an garch modelling??? , because sometimes ,like the 40% , when i did simulate garch procces ( i made like 1200 obs) , the kurtosis is nongreater to 3??
        atte a ustedes Ricardo Zambrano Aguilera

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Reply | Threaded
Open this post in threaded view
|

Re: some doubts in garch models

Patrick Burns-2
Are you sure that you are getting kurtosis and not
excess kurtosis?  I've never seen market data that
weren't fat tailed.  If you have an example, I'm keen
to see it.

Using 'qqnorm' should give you a good visual clue.

Patrick Burns
[hidden email]
+44 (0)20 8525 0696
http://www.burns-stat.com
(home of S Poetry and "A Guide for the Unwilling S User")

Ricardo Zambrano Aguilera wrote:

> Thank a lot mr Diethelm Wuertz and mr Patrick Burns my best regards..., but then i have another doubt... and it is a big
> if my data... the exchange rate (dollar (USA) - peso(Chile)), between 1-1-2001 - 12/12/2005, (i´m working with the log-returns), my first intention is to aplly an ARMA -GARCH , but  their Kurtosis is 1.182 < 3,  ¿ can i use an garch modelling??? , because sometimes ,like the 40% , when i did simulate garch procces ( i made like 1200 obs) , the kurtosis is nongreater to 3??
> atte a ustedes Ricardo Zambrano Aguilera
>
>_______________________________________________
>[hidden email] mailing list
>https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>
>
>
>  
>

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Reply | Threaded
Open this post in threaded view
|

Re: some doubts in garch models

Patrick Burns-2
In reply to this post by Ricardo Zambrano Aguilera
Your Kurtosis is actually excess kurtosis:

 > mean(((dolpeso - mean(dolpeso))/sd(dolpeso))^4)
[1] 4.182083

(Ricardo sent me the data privately.)

Patrick Burns
[hidden email]
+44 (0)20 8525 0696
http://www.burns-stat.com
(home of S Poetry and "A Guide for the Unwilling S User")

Ricardo Zambrano Aguilera wrote:

> Thank a lot mr Diethelm Wuertz and mr Patrick Burns my best regards..., but then i have another doubt... and it is a big
> if my data... the exchange rate (dollar (USA) - peso(Chile)), between 1-1-2001 - 12/12/2005, (i´m working with the log-returns), my first intention is to aplly an ARMA -GARCH , but  their Kurtosis is 1.182 < 3,  ¿ can i use an garch modelling??? , because sometimes ,like the 40% , when i did simulate garch procces ( i made like 1200 obs) , the kurtosis is nongreater to 3??
> atte a ustedes Ricardo Zambrano Aguilera
>
>_______________________________________________
>[hidden email] mailing list
>https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>
>
>
>  
>

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Reply | Threaded
Open this post in threaded view
|

Re: [QUAR] Re: some doubts in garch models

Tim Hesterberg
In reply to this post by Patrick Burns-2
Note that different people use different terminology.

The terminology I see most is:
        kurtosis = E((X-mu)^4) / var(X)^2 - 3
With this terminology, kurtosis > 0 indicates longer tails than Gaussian.
The S-PLUS kurtosis function computes this kurtosis.

Alternate terminology:
        excess kurtosis = E((X-mu)^4) / var(X)^2 - 3
        kurtosis = E((X-mu)^4) / var(X)^2

Tim Hesterberg

========================================================
| Tim Hesterberg       Research Scientist              |
| [hidden email]  Insightful Corp.                |
| (206)802-2319        1700 Westlake Ave. N, Suite 500 |
| (206)283-8691 (fax)  Seattle, WA 98109-3012, U.S.A.  |
|                      www.insightful.com/Hesterberg   |
========================================================

>Are you sure that you are getting kurtosis and not
>excess kurtosis?  I've never seen market data that
>weren't fat tailed.  If you have an example, I'm keen
>to see it.
>
>Using 'qqnorm' should give you a good visual clue.
>
>Patrick Burns
>[hidden email]
>+44 (0)20 8525 0696
>http://www.burns-stat.com
>(home of S Poetry and "A Guide for the Unwilling S User")
>
>Ricardo Zambrano Aguilera wrote:
>
>> Thank a lot mr Diethelm Wuertz and mr Patrick Burns my best regards..., but then i have another doubt... and it is a big
>> if my data... the exchange rate (dollar (USA) - peso(Chile)), between 1-1-2001 - 12/12/2005, (i´m working with the log-returns), my first intention is to aplly an ARMA -GARCH , but  their Kurtosis is 1.182 < 3,  ¿ can i use an garch modelling??? , because sometimes ,like the 40% , when i did simulate garch procces ( i made like 1200 obs) , the kurtosis is nongreater to 3??
>> atte a ustedes Ricardo Zambrano Aguilera

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance